Depression is a significant symptom for approximately one in four palliative care patients. This study investigates the performance of three screening tools. Patients were asked to verbally rate their mood on a scale of 0-10; to respond "yes" or "no" to the question "Are you depressed?," and to complete the Edinburgh depression scale. They were also interviewed using a semi-structured clinical interview according to DSM-IV criteria. Complete data was available for 74 patients. For the single question, a "yes" answer had a sensitivity of 55% and specificity 74%. The Edinburgh depression scale at a cut-off point of > or =13 had a sensitivity of 70% and specificity of 80%. The verbal mood item with a cut-off point of > or =3 had a sensitivity of 80% and specificity of 43%. The Edinburgh depression scale proved to be the most reliable instrument for detecting clinical depression in palliative care patients.